Maddison Project Database 2020

ggplot2 plotly

We will work with the data provided by the Maddison Project and create a graph showing the GDP per capita growth of selected countries.

Francisco Guzmán
11-15-2023

Libraries

library(readxl)
  library(ggplot2)
  library(plotly)
  library(Cairo)
  library(scales)
  library(ggthemes)
  library(dplyr)

Database

Import

We obtain the data from the Maddison Project Database 2020. Then, we import it.

mpd2020 <- as.data.frame(read_excel("mpd2020.xlsx", sheet = "Full data"))

unique()

The following command might be useful for country selection. In this case, we’ll compare Chile’s growth with Norway, Sweden, Finland, New Zealand, and Australia.

unique(mpd2020[c("countrycode", "country")])

summary()

With the following command, we are interested in seeing the period that will be covered in the graph, considering the available data for Chile.

mpd2020 %>% 
  filter(countrycode == "CHL") %>%
  summary()

We see that there is data for Chile from 1800, so the graph will start from that year.

filter()

Given the information from the previous two steps, we filter the database to leave it with what we need. We can create a new one in case we need the complete database for something else.

df_graf1 <- mpd2020 %>%
  filter(countrycode %in% c("CHL", "NOR", "SWE", "FIN", "NZL", "AUS"),
         year>=1800)

Graph

ggplot()

my_palette = c("#DA0000", "#239f40", "#C4961A", "#CC79A7", "#C3D7A4", "#293352")

g1 <- ggplot(data = df_graf1, aes(year, gdppc, color=country)) +
  geom_line(size = 1) +
  labs(title = "International Comparison of Real GDP per capita in 2011$",
       subtitle = "1800-2018",
       caption = "Source: Personal elaboration with Maddison Project Database 2020",
       x = "",
       y = "",
       col = NULL) +
  scale_color_manual(values = my_palette) +
  scale_x_continuous(breaks=seq(1800,2020, by = 10), position = "bottom") +
  scale_y_continuous(labels = dollar) +
  theme_tufte() +
  theme(plot.title = element_text(size = rel(1.4)),
        plot.subtitle = element_text(size = rel(1.2), margin=margin(0,0,20,0)),
        plot.caption = element_text(hjust = 0, face= "italic"),
        axis.text.x = element_text(angle = 90, vjust=.5, size = 10),
        axis.text.y = element_text(angle = 0, vjust=0, size = 10),
        axis.ticks.length.x = unit(.25, "cm"),
        axis.ticks.x = element_line(size = .5),
        axis.ticks.y = element_blank(),
        panel.grid.major.y = element_line(colour = "grey90"),
        panel.grid.minor.y = element_line(colour = "grey90"))

plotly()

With this command, we get an interactive plot.

fig <- ggplotly(g1)
fig

Code

Show code 
library(readxl)
library(ggplot2)
library(plotly)
library(Cairo)
library(scales)
library(ggthemes)
library(dplyr)

mpd2020 <- as.data.frame(read_excel("mpd2020.xlsx", sheet = "Full data"))

unique(mpd2020[c("countrycode", "country")])

mpd2020 %>% 
  filter(countrycode == "CHL") %>%
  summary()
df_graf1 <- mpd2020 %>%
  filter(countrycode %in% c("CHL", "NOR", "SWE", "FIN", "NZL", "AUS"),
         year>=1800)

my_palette = c("#DA0000", "#239f40", "#C4961A", "#CC79A7", "#C3D7A4", "#293352")

g1 <- ggplot(data = df_graf1, aes(year, gdppc, color=country)) +
  geom_line(size = 1) +
  labs(title = "International Comparison of Real GDP per capita in 2011$",
       subtitle = "1800-2018",
       caption = "Source: Personal elaboration with Maddison Project Database 2020",
       x = "",
       y = "",
       col = NULL) +
  scale_color_manual(values = my_palette) +
  scale_x_continuous(breaks=seq(1800,2020, by = 10), position = "bottom") +
  scale_y_continuous(labels = dollar) +
  theme_tufte() +
  theme(plot.title = element_text(size = rel(1.4)),
        plot.subtitle = element_text(size = rel(1.2), margin=margin(0,0,20,0)),
        plot.caption = element_text(hjust = 0, face= "italic"),
        axis.text.x = element_text(angle = 90, vjust=.5, size = 10),
        axis.text.y = element_text(angle = 0, vjust=0, size = 10),
        axis.ticks.length.x = unit(.25, "cm"),
        axis.ticks.x = element_line(size = .5),
        axis.ticks.y = element_blank(),
        panel.grid.major.y = element_line(colour = "grey90"),
        panel.grid.minor.y = element_line(colour = "grey90"))

# To save in png format
ggsave(g1, filename = "plot.png", type = 'cairo', dpi = 300,
       width = 18, height = 13.5, units = "cm")

# For higher quality (pdf or eps)
ggsave(g1, filename = "plot.pdf", device = cairo_pdf, dpi = 300,
       width = 18, height = 13.5, units = "cm")

ggsave(g1, filename = "plot.eps", device = cairo_ps, dpi = 300,
       width = 18, height = 13.5, units = "cm")
knitr::include_graphics("plot.png")

fig <- ggplotly(g1)
fig

References